Is there a tutorial on setting up fuel tanks and pumps?

[Guest sphing]

I've tried changing values in the .air file and the SET_FUEL_TRANSFER_FORWARD command does not want to work.I'm triggering the command from within my own XML gauge file.Center2 is the aux. tank and center is the wing center tank. I want the aux. tank to transfer fuel when the center wing is below 900Kg.Is the concord the only plane that has ever used the SET_FUEL_TRANSFER_FORWARD and actually had it work?Also, the FUEL_SELECTOR_SET commands do not allow me to set CENTER2 or CENTER3 ...only CENTER. If I set to ALL, I will see fuel being drawn from the CENTER2 tank and the LEFT MAIN and RIGHT MAIN but not CENTER. What the heck is going on? The default logic is hard to understand.Any further recommendations on getting the CENTER2 to CENTER fuel transfer working using the SET_FUEL_TRANSFER_FORWARD command?Thanks,John

[Guest Ron Freimuth]

>Ron,>>Boeing 767-300ER.>I tried different A:Var's, but sometimes Fueltemperature is>below freezing point, -37C (Gives Eng Failure etc.), not so>real.>So i use now (Ambient temperature - Altitude/1500)>At N 80 minimum measured temp. was -26C.>..>May be he reads this thread?!>Jan I'm expecting to make a complex FTT for the DF 727. I sorta figured it out months ago. I'll consider specific heat of jet fuel and the tank, thermal resistances, effect of airflow on cooling (also, on heating at higher Mach speeds), ...... etc. I ripped out the sun_angle code posted here and expect I will take COS(sun_angle) to account for unclouded sunshine. After all, a car sitting in the sun really gets hot. So would a parked AC. And, one doesn't want to exceed the flash point of JP-4. As the sun sets, the fuel will slowly cool. But I can't directly get at the clouds that might shield the sun. I think WX and percent cloud cover can be gotten through FSUIPC, but DF AC don't depend on FSUIPC. I'd like to make the fuel gel if the tank heaters aren't turned on when appropriate. Perhaps cut off fuel flow - though toggling it on and off rapidly might emulate just restricted flow. I'll try to come up with a general form for fuel temperature. Then adjust the constants to make it change in what seems an appropirate way. I'm sure it takes a long time to cool or heat up to dangerous levels. Say a couple of hours. Or, rather than doing a 'physics model', I may take a more intuitive approach and use the 'Time Constant' code I put in an XML file. That can slow any variable one puts into it. Further, I'll only strobe the associated code once a minute or so so it adds neglible CPU loading. I worked that out also in the test gauge I've been working on. Incidently, I think I'll use (K1*OAT + (1-K1)*TAT) to get a weighted average of the two. TAT is typically 20 C higher than OAT, but mainly heats the LE of the wing. I also want air flow to cool down a hot wing. Like the fan on your CPU heatsink. Anyone wanna record the fuel temperature in his auto? Be sure to leave it sitting in the sun, then record the temperature as you drive it. Drive up a good mountain to drop the ambient temperature and see how far the fuel temp lags the OAT. ;) Ron

[Guest Ron Freimuth]

>I've tried changing values in the .air file and the>SET_FUEL_TRANSFER_FORWARD command does not want to work.>>I'm triggering the command from within my own XML gauge file. That command may need the Concorde Transfer Pump. Incidently, I think it works off the CG limits set in aircraft.cfg when it is on Auto. But, I would think it also gets some variable from the intrinsic Concorde code in sim1.dll so it knows the CG to aim for im pumping fule. OTOH, in manual mode it just moves fuel between Center 1 and Center 3. >Center2 is the aux. tank and center is the wing center tank. I>want the aux. tank to transfer fuel when the center wing is>below 900Kg. It may only work between Center1 and Center3. You might be able to put Center3 where the aux tank is.>Any further recommendations on getting the CENTER2 to CENTER>fuel transfer working using the SET_FUEL_TRANSFER_FORWARD>command?>Thanks,>John Seems it could be done by resetting fuel levels in each tank every second or so; on up, the other down. But, it's been said you can't add fuel to a tank. Ron

[spokes2112 179]

John,1. Make sure you have the tanks properly set in the aircraft.cfg.Center1= 0, 0, 0, 450.000, 0.000Center2= 0, 0, 0, 225.000, 0.000Those two entries are the ones needed for the "Concorde" type transfer.1A..May also want to make sure this is set in the ELECTRICAL section of the aircraft.cfg.fuel_transfer_pump = 0, 0 , 1.0fuel_pump = 0, 0 , 1.02. These are the commands for doing transfer....(>K:SET_FUEL_TRANSFER_AFT) center1 -> center2(>K:SET_FUEL_TRANSFER_AUTO) ***** This relys on CG of Concorde, NOT recommended. (>K:SET_FUEL_TRANSFER_FORWARD) center2 -> center1(>K:SET_FUEL_TRANSFER_OFF) 3. Theses are the enums for setting fuel tanks -0 = OFF1 = ALL2 = LEFT3 = RIGHT4 = LT AUX5 = RT AUX6 = CENTER17 = CENTER28 = CENTER3 ??? = LT TIP ??? = RT TIP ??? = EXT 1 ??? = EXT2 ?? 14 = CRFD L -> R15 = CRFD R -> L""Any further recommendations on getting the CENTER2 to CENTER fuel transfer working using the SET_FUEL_TRANSFER_FORWARD command?""See the coding in the last Fuel Test .zip I posted, some of the coding may be helpful.http://forums.avsim.net/dcboard.php?az=vie...t&file_id=29064The Aircraft Container SDK ( for 02 ) may help also in the assignment of tanks to the aircraft.http://zone.msn.com/flightsim/FS02DevDeskSDK02.aspFrom testing it works as described..Regards,Roman

[Guest]

Not sure if I should put this in this already oversized thread, but...I'm not familiar with the fuelsystem of a real DC3, but I find it odd that the left engine draw fuel from both mains after the right has been shut down (after a simulated failure, Vmc speeds/dynamics suck btw), even though the fuel selector had the left main selected! Bug still persist?Oh, and by default when you load an aircraft, all fuel selectors are set to ALL tanks :(

[phjvh 0]

Karl,With "my logic" Fuel Selectors are set accordingly, i.e. the Center Tank first(!?).(Checked this by Reload Panels, another AC and restarting the game, in flight or "cold and dark")Logic:All (8, (G:Var1-8) 0 ==) pumps on (0!) then: 0 (>G:Var9) els 1.If 0, as always at any restart, the logic commands Fuelselectors set on the Center Tank, dependant of the amount of fuel (> 1.5%), as in the real world; if < 1.5%, left engine is feed by the left- and right engine by the right main tank.By imbalance > 1.5% between left and right, the fullest tank is emptied first of course, all automatic.If total fuel drops below 1.5% Selectors are set on ALL.(All controlled by 's.)Pressing one of the pumps sets (G:Var9) to 1, where after i can choose any tank, using all the possible variations of the pumpswitches ("fuelpumps") and x-feed valves(2) as fuelselectors.The "fwd and aft fuelpumps" of the right- and left main tanks contain the 4 Elec Fuel pumps.1 and 2 give the fuel pressure for starting the engines.3, altough not necessary, "provides" fuel for the APU, using the (A:Var) Eng3 Elec Fuel pump switch, otherwise no start...In contrast to others i see a good x-feed system, using mains instead of externals.I mean p.e.:Left main is emptying when both selectors are on the left and the 2 engines are drawing fuel from her vice versa.Level of right- and center tanks remain steady.Or is this not what you mean by crossfeed?Jan"Procul Negotiis"

[Guest Ron Freimuth]

>John,>>1. Make sure you have the tanks properly set in the>aircraft.cfg.>>Center1= 0, 0, 0, 450.000, 0.000>Center2= 0, 0, 0, 225.000, 0.000>>Those two entries are the ones needed for the "Concorde" type>transfer................>Regards,>Roman Ah, they are Center1 and Center2. How did I figure Center3? --------------------------------- It appears the Doublas DC-3 in FS9 engines can be independently set to Left or Right tanks. I set Left Eng to R Tank and vica versa but didn't try to see if both on Left used only fuel from the Left tank. I checked the XML code for this, it is in the panelXXX.CAB file. It was short, but I wasn't sure of just how it worked. It looked like the same code was used for the Ford Tri-motor. Ron

[Guest sphing]

Ron,I don't have FS9.Could you zip the two .CAB files with the XML and post it here?Thanks,John

[Guest]

Maybe my problems has been because I've had aux and tiptanks as well as mains (although using externals for these), yet no center tanks. See the attached files for my setup that worked for FS2004, things _might_ behave better now with FS9 though.attached files; part of aircraft.cfg describing the fuel system in pretty good detail, and the selector code that works and how to make it not work... With FS2002 that is, can't check yet with FS9.

[Guest Ron Freimuth]

>center tanks. See the attached files for my setup that worked>for FS2004, things _might_ behave better now with FS9 though. Saved.>attached files; part of aircraft.cfg describing the fuel>system in pretty good detail, and the selector code that works>and how to make it not work... With FS2002 that is, can't>check yet with FS9. Could you explain the RPN stack operations in this string:"A:Fuel tank selector 2,enum) sp0 l0 0(*Off*)" I have assumed 'sp0' stores st in '0', then pops it off the stack. But that doesn't make sense above. Since st would be stored in '0', popped off, then reloaded with the 'l0'. I think it was then compared with the '0' in what followed. Has anyone found the depth of the stack? I haven't worried about it overflowing, but perhaps I should pop off variables if they are no longer needed. I found out the hard that the 8087 NPX stack didn't overlow nicely, as it does in an HP RPN calculator. Ron

[spokes2112 179]

Ron, I am not "programmer savy" but here goes... I have yet to figure out the pop.. Maybe only necessary if a R is used. (reverse) The S vars will always get set to zero after each read of the gauge, not sure what happens to the stack itself ?? Perhaps pointers get set to non volatile vars IE G:Vars and L:Vars and the stack is discarded ???.. Never used sp yet..Regards,RomanGREEN BAY PACKERS][/i :-ukliam :-beerchug

[Guest Ron Freimuth]

>Ron,>> I am not "programmer savy" but here goes... I have yet to>figure out the pop.. Maybe only necessary if a R is used.>(reverse) Your quotes are messed up since I checked the 'plain text' box to preserve ">". Having programmed the 8087, I guessed that 'sp0' also removes the variable from the stack. While 's0' doesn't. In the 80X87 NPX, 'FSTOP my_var' puts the 'top' of the eight register 80X87 stack, ST(0) into 'my_var'; then pops ST(0) off the stack (stack variables move up by one). While 'FSTO my_var' leaves ST(0) register the same. It would be easy to test, but awaits more important XML programming. At least I'm getting so I don't make all the errors I did in the past. > The S vars will always get set to zero after each>read of the gauge, not sure what happens to the stack itself>?? Perhaps pointers get set to non volatile vars IE G:Vars and>L:Vars and the stack is discarded ???>Roman I do know if one a (>L:var,number) is in the code, then the stack top is popped. I do a lot of "s0"'s before such elements, then a "l0" to recover the value to the stack. I figure that's better than reloading (L:var,number), etc. for the next operation that still needs the result. But, don't want to depend on a "l0" for long since it's too difficult to keep track of the associated "0, 1, ..." registers. I've used "S1", "S2" a few times when I wanted to hold intermediate calculations for later loading to st (stack top; some would say bottom).RonPS: That 'PE editor' also will disassemble SIM1.DLL, etc. The disassembly listing is mostly long rows of things like "fmult [dword ptr] 01234". I noted some of the accesses were of variables associated with strings in aircraft.cfg. I expect a lot of work would add to understanding of sim1.dll -- also panel.dll, etc.

[Guest]

Hmm, why do I use sp0 l0? I have no idea really, maybe just to get equal code for all possible variants. The selector code was based on the baron code for FS2004, maybe that used sp0 l0 as well - it does now anyway (Engine Controls Unit.xml, the new gps'es has this also). Although now it uses some new macro elements which appear to be new. Hmm, and I thought nothing new had happened with the Baron gauges since the icing switches are still fake.

[Guest sphing]

Roman, Karl or Ron,I've read and re-read this thread to understand the FS fuel system better. It's a wealth of info, but I'm still unclear about a few things. Primarily, the SET_FUEL_TRANSFER... series of commands. It seems that I should be able to get this to work between center1 and center2 tanks. It was suggested to also play with the .air file and set some variables to match the "concord" .air file. (namely: electric_pumps=1) Also it was recommended that the [electrical] section has the fuel_transfer_pump= and fuel_pump= set.I not getting something right because the trasnfer does not work for me.There has been some back and forth between using center1 and center3 vs. center1 and center2. also, the fuel cross-feed issue that should be corrected by using external1 and external2 vs. rightmain and leftmain.I know you have already gone over this quite extensively for my benefit, but would you please try explaining it one more time? May summarize this giant thread into 1 or 2 posts?Thanks,John

[spokes2112 179]

John, How about posting or attaching the code snippet and/or the Fuel section of aircraft.cfg too ?? That way we can take a peek at whats up ...Yes this is a long thread --- A battle between the FDE gurus and XML guys for the longest one.. LOL !!!! Good reading though..Regards,RomanGREEN BAY PACKERS][/i :-ukliam :-beerchug